Skip to content

Latest commit

 

History

History
108 lines (69 loc) · 4.24 KB

README.md

File metadata and controls

108 lines (69 loc) · 4.24 KB
Raw

NOTE: FreeNAS/TrueNAS now has temperature monitoring built-in to the reporting dashboard. I'll leave this tool up for anyone who's interested and will accept pull requests, but it will be fairly unmaintained.

freenas-temperature-graphing

Bash scripts to graph FreeNAS CPU and drive temperatures. Graphs are generated in the same directory as the scripts (not the web interface). Tested on 9.3, 9.10, and 11.2.

Example output:

CPU temperatures per minute

Drive temperatures per minute

CPU temperatures per 5 minutes

Drive temperatures per 5 minutes

 

Installation:

  1. Copy the .sh files to a share on your server. Take note of the full path to the files (ex. /mnt/mainpool/myshare/temperature-monitoring/rrd.sh)

  2. Schedule the scripts to run regularly. The recommended method is via the FreeNAS Tasks panel in the web interface, since any cronjobs in crontabs are wiped out during upgrades. You need one task for data collection, and another task for data graphing:

    1. Create a new task (see tasks screenshot)

    2. Set the user to root (SMART data access requires super-user privs). See screenshot 1.

    3. Enter the path to the script run, followed by the path to the data file (the rrd.sh script will create it if it doesn't exist yet). See screenshot 1. For example, if you want temperatures to be collected every 5 minutes you would create two tasks:

      1. The data collection task would have the "Command" field set to: /mnt/mainpool/misc/temperature-monitoring/rrd.sh /mnt/mainpool/misc/temperature-monitoring/temps-5min.rrd

      2. The graphing task would have the "Command" field set to: /mnt/mainpool/misc/temperature-monitoring/rrd-graph.sh /mnt/mainpool/misc/temperature-monitoring/temps-5min.rrd

      3. NOTE: The time part of the name (5min) is important, since the scripts use it to know how to format the data. It has to be a number followed by "min".

    4. Enter the time between runs: minutes (5), hours (1), and days (1). Set it to run every month and day of the week. See screenshot 1 and screenshot 2

    5. Redirect Stdout (otherwise you'll get emails every time the graphs are generated). Make sure the task is enabled. See screenshot 3

    6. After about 5 intervals, you should have enough data that the graphs will start contain visible data. The graphs will appear as png files in the same directory as the scripts (regardless of where the rrd files are stored).

 

Extras:

The rrd-graph.sh script contains code to graph each device individually as well as all together. Just uncomment the function call at the bottom of rrd-graph.sh.

The colors can be adjusted at the top of the rrd-graph.sh file.

The temperature lines can be changed in the rrd-graph.sh file, or by setting enivronment variables when running: Example: MAXGRAPHTEMP=55 /pathto/rrd-graph.sh /pathto/temps-5min.rrd Temperature variables are: MAXGRAPHTEMP MINGRAPHTEMP SAFETEMPMAX SAFETEMPMIN

Troubleshooting:

  • Try running the script manually with the verbose flag (-v). Example: /mnt/mainpool/misc/temperature-monitoring/rrd-graph.sh -v /mnt/mainpool/misc/temperature-monitoring/temps-5min.rrd

  • Errors and possible causes:

    • "ERROR: mmaping file '/mnt/mainpool/misc/temperature-monitoring/temps-60min.rrd': Invalid argument"

      • This probably means that the rrd file wasn't initialized properly. Try deleting it and running the script again (it should initialize a new one if it's missing)

    • ERROR: opening '/mnt/mainpool/misc/temperature-monitoring/temps-60min.rrd': No such file or directory

      • This may happen on the first run, if the graphing script runs before the temperature-gathering script runs. It should only happen once.

Pull requests welcome